Arrays and matrices

Matrices are at the heart of most scientific Python and artificial intelligence libraries because they are very convenient for storing a lot of related data. They are also suitable for really fast bulk processing, and calculations on them can be performed much faster than you could achieve with many separate variables. In some cases, these calculations can even be offloaded to the GPU for even faster processing.

Note that a 0D matrix is effectively a single number, a 1D matrix is a regular array, and there is no real limit to the number of dimensions you can use. It should be noted that both size and processing time quickly increase with multiple dimensions, of course.

NumPy – Fast arrays and matrices

The numpy package spawned most of the scientific Python development and is still used at the core of many of the libraries covered in this chapter and the next. The library is largely (where it matters, at least) written in C, which makes it extremely...